Kokkos Core Kernels Package Version of the Day
Class List
Here are the classes, structs, unions and interfaces with brief descriptions:
Kokkos::Sequential::AddMatrixEntries< ScalarType >Compute C_ij = alpha * A_ij + beta * B_ij
Kokkos::Details::ArithTraits< T >Traits class for arithmetic on type T
Kokkos::Bitset< Device >A thread safe view to a bitset
Kokkos::ConstBitset< Device >
Kokkos::CrsArray< DataType, Arg1Type, Arg2Type, SizeType >Compressed row storage array
Kokkos::CrsMatrix< ScalarType, OrdinalType, Device, MemoryTraits, SizeType >Compressed sparse row implementation of a sparse matrix
Kokkos::CudaSpaceCuda memory management
Kokkos::Impl::FunctorPolicyExecutionSpace< Functor, Policy, EnableFunctor, EnablePolicy >Given a Functor and Execution Policy query an execution space
Kokkos::Future< Arg1, Arg2 >
Kokkos::HostSpaceMemory management on the host for devices
Kokkos::Details::InnerProductSpaceTraits< T >Traits class for inner product space operations on type T
Kokkos::LayoutLeftMemory layout tag indicating left-to-right (Fortran scheme) striding of multi-indices
Kokkos::LayoutRightMemory layout tag indicating right-to-left (C or lexigraphical scheme) striding of multi-indices
Kokkos::LayoutStrideMemory layout tag indicated arbitrarily strided multi-index mapping into contiguous memory
Kokkos::LayoutTileLeft< ArgN0, ArgN1, IsPowerOfTwo >Memory layout tag indicating left-to-right (Fortran scheme) striding of multi-indices by tiles
Kokkos::TeamPolicy< ExecSpace, WorkArgTag >::member_typeParallel execution of a functor calls the functor once with each member of the execution policy
Kokkos::pair< T1, T2 >Replacement for std::pair that works on CUDA devices
ParallelForImplementation of the ParallelFor operator that has a partial specialization for the device
ParallelReduceImplementation detail of parallel_reduce
ParallelScanImplementation detail of parallel_scan
Kokkos::QthreadExecution space supported by Qthread
Kokkos::RangePolicy< ExecSpace, WorkArgTag, IntType, GranularityPowerOfTwo >Execution policy for work over a range of an integral type
Kokkos::Impl::ReduceAdapter< FunctorType, ScalarType >Implementation detail of parallel_reduce
Kokkos::SerialKokkos device for non-parallel execution
Kokkos::SparseRowView< MatrixType >View of a row of a sparse matrix
Kokkos::SparseRowViewConst< MatrixType >Const view of a row of a sparse matrix
Kokkos::StaticCrsGraph< DataType, Arg1Type, Arg2Type, SizeType >Compressed row storage array
Kokkos::TeamPolicy< ExecSpace, WorkArgTag >Execution policy for parallel work over a league of teams of threads
Kokkos::UnorderedMap< Key, Value, Device, Hasher, EqualTo >Thread-safe, performance-portable lookup table
Kokkos::UnorderedMapInsertResultFirst element of the return value of UnorderedMap::insert()
Kokkos::Impl::VerifyExecutionCanAccessMemorySpace< Kokkos::CudaSpace, Kokkos::HostSpace >
Kokkos::Impl::VerifyExecutionCanAccessMemorySpace< Kokkos::HostSpace, Kokkos::CudaSpace >
ViewView to an array of data
Kokkos::Impl::ViewDataHandle< ViewTraits, Enable >ViewDataHandle provides the type of the 'data handle' which the view uses to access data with the [] operator. It also provides an allocate function and a function to extract a raw ptr from the data handle. ViewDataHandle also defines an enum ReferenceAble which specifies whether references/pointers to elements can be taken and a 'return_type' which is what the view operators will give back. Specialisation of this object allows three things depending on ViewTraits and compiler options: (i) Use special allocator (e.g. huge pages/small pages and pinned memory) (ii) Use special data handle type (e.g. add Cuda Texture Object) (iii) Use special access intrinsics (e.g. texture fetch and non-caching loads)
Kokkos::Impl::ViewSpecialize< ValueType, void, LayoutLeft, MemorySpace, MemoryTraits >Default view specialization has LayoutLeft, LayoutRight, or LayoutStride
Kokkos::ViewTraits< DataType, Arg1, Arg2, Arg3 >Traits class for accessing attributes of a View
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Friends